An intriguing aspect of prostate cancer (CaP) skeletal metastases is that they form primarily osteoblastic lesions. The osteoblastic nature of CaP metastases may provide clues to their biology; however, the mechanisms of bone production at CaP skeletal metastatic sites are poorly understood. Potential mediators of this effect are bone morphogenetic proteins (BMP), which are proteins that induce osteoblastogenesis. We present data that CaP tumor metastases produce BMPs in situ and that BMP-7 expression is dysregulated in osteoblastic CaP cell lines. We hypothesize that as CaP cells progress to a more aggressive state, BMP expression is dysregulated at the transcriptional level and the resulting overexpression of BMP promotes the development of osteoblastic lesions. To test our hypothesis, we will perform the following specific aims: Aim 1: Determine the extent of BMPs contribution to the mechanism of CaP-induced bone formation. In this aim, we will implant osteoblastic CaP tumors into human fetal bone implanted in SCID mice (SCID-hu) and will then determine the effect of (1) inhibiting BMP (antibodies and antisense) or (2) overexpressing BMP in a low BMP-expressing CaP cell line on osteoblastic lesion formation. Aim 2: Investigate the etiology of dysregulated BMP-7 expression in CaP cells. To perform this, we will identify cis-acting sites and trans-acting factors that account for the differential regulation of the BMP-7 promoter in osteoblast-inducing C4-2B CaP cells compared to its non-osteoblast-inducing parental LNCaP cells. We will also test this in vivo using real-time imaging of the promoter activation. Aim 3: Determine if BMP promotion of osteoblastic activity in CaP metastases depends on vascular endothelial growth factor (VEGF). We have data showing that BMPs increase VEGF expression and that inhibition of VEGF diminishes CaP-induced pro-osteoblast activity. In this aim, we will test the ability of BMPs to regulate VEGF expression and activity (including angiogenesis, as well as pro-osteoblastic activity) by blocking BMP activity. We will also determine the effects of blocking VEGF on CaP-mediated pro-osteoblastic activity. When completed, this project clues that may enable design of strategies to prevent or delay progression of CaP skeletal metastases, which are common and painful sequelae of prostate cancer.